Satish Mahadevan Srinivasan

LocSigDB: a database of protein localization signals

LocSigDB (http://genome.unmc.edu/LocSigDB/) is a manually curated database of experimental protein localization signals for eight distinct subcellular locations; primarily in a eukaryotic cell with brief coverage of bacterial proteins. Proteins must be localized at their appropriate subcellular compartment to perform their desired function. Mislocalization of proteins to unintended locations is a causative factor for many human diseases; therefore, collection of known sorting signals will help support many important areas of biomedical research. By performing an extensive literature study, we compiled a collection of 533 experimentally determined localization signals, along with the proteins that harbor such signals. Each signal in the LocSigDB is annotated with its localization, source, PubMed references and is linked to the proteins in UniProt database along with the organism information that contain the same amino acid pattern as the given signal. From LocSigDB webserver, users can download the whole database or browse/search for data using an intuitive query interface. To date, LocSigDB is the most comprehensive compendium of protein localization signals for eight distinct subcellular locations. Database URL: http://genome.unmc.edu/LocSigDB/

MetaID: A novel method for identification and quantification of metagenomic samples

BackgroundAdvances in next-generation sequencing (NGS) technology has provided us with an opportunity to analyze and evaluate the rich microbial communities present in all natural environments. The shorter reads obtained from the shortgun technology has paved the way for determining the taxonomic profile of a community by simply aligning the reads against the available reference genomes. While several computational methods are available for taxonomic profiling at the genus- and species-level, none of these methods are effective at the strain-level identification due to the increasing difficulty in detecting variation at that level. Here, we present MetaID, an alignment-free n-gram based approach that can accurately identify microorganisms at the strain level and estimate the abundance of each organism in a sample, given a metagenomic sequencing dataset.ResultsMetaID is an n-gram based method that calculates the profile of unique and common n-grams from the dataset of 2,031 prokaryotic genomes and assigns weights to each n-gram using a scoring function. This scoring function assigns higher weightage to the n-grams that appear in fewer genomes and vice versa; thus, allows for effective use of both unique and common n-grams for species identification. Our 10-fold cross-validation results on a simulated dataset show a remarkable accuracy of 99.7% at the strain-level identification of the organisms in gut microbiome. We also demonstrated that our model shows impressive performance even by using only 25% or 50% of the genome sequences for modeling. In addition to identification of the species, our method can also estimate the relative abundance of each species in the simulated metagenomic samples. The generic approach employed in this method can be applied for accurate identification of a wide variety of microbial species (viruses, prokaryotes and eukaryotes) present in any environmental sample.ConclusionsThe proposed scoring function and approach is able to accurately identify and estimate the entire taxa in any metagenomic community. The weights assigned to the common n-grams by our scoring function are precisely calibrated to match the reads up to the strain level. Our multipronged validation tests demonstrate that MetaID is sufficiently robust to accurately identify and estimate the abundance of each taxon in any natural environment even when using incomplete or partially sequenced genomes.

Mining for class-specific motifs in protein sequence classification

BackgroundIn protein sequence classification, identification of the sequence motifs or n-grams that can precisely discriminate between classes is a more interesting scientific question than the classification itself. A number of classification methods aim at accurate classification but fail to explain which sequence features indeed contribute to the accuracy. We hypothesize that sequences in lower denominations (n-grams) can be used to explore the sequence landscape and to identify class-specific motifs that discriminate between classes during classification. Discriminative n-grams are short peptide sequences that are highly frequent in one class but are either minimally present or absent in other classes. In this study, we present a new substitution-based scoring function for identifying discriminative n-grams that are highly specific to a class.ResultsWe present a scoring function based on discriminative n-grams that can effectively discriminate between classes. The scoring function, initially, harvests the entire set of 4- to 8-grams from the protein sequences of different classes in the dataset. Similar n-grams of the same size are combined to form new n- grams, where the similarity is defined by positive amino acid substitution scores in the BLOSUM62 matrix. Substitution has resulted in a large increase in the number of discriminatory n-grams harvested. Due to the unbalanced nature of the dataset, the frequencies of the n-grams are normalized using a dampening factor, which gives more weightage to the n-grams that appear in fewer classes and vice-versa. After the n-grams are normalized, the scoring function identifies discriminative 4- to 8-grams for each class that are frequent enough to be above a selection threshold. By mapping these discriminative n-grams back to the protein sequences, we obtained contiguous n-grams that represent short class-specific motifs in protein sequences. Our method fared well compared to an existing motif finding method known as Wordspy. We have validated our enriched set of class-specific motifs against the functionally important motifs obtained from the NLSdb, Prosite and ELM databases. We demonstrate that this method is very generic; thus can be widely applied to detect class-specific motifs in many protein sequence classification tasks.ConclusionThe proposed scoring function and methodology is able to identify class-specific motifs using discriminative n-grams derived from the protein sequences. The implementation of amino acid substitution scores for similarity detection, and the dampening factor to normalize the unbalanced datasets have significant effect on the performance of the scoring function. Our multipronged validation tests demonstrate that this method can detect class-specific motifs from a wide variety of protein sequence classes with a potential application to detecting proteome-specific motifs of different organisms.

Data aggregation in partially connected networks

With the diverse new capabilities that sensor and ad hoc networks can provide, applicability of data aggregation is growing. Data aggregation is useful in dealing with multi-value domain information, which often requires approximate agreement decisions among nodes. In contrast to fully connected networks, the research on data aggregation for partially connected networks is very limited. This is due to the complexity of formal proofs and the fact that a node may not have a global view of the entire network, which makes it difficult to attain the convergence properties. The complexity of the problem is compounded in the presence of message dropouts, faults, and orchestrated attacks. By exploiting the properties of Discrete Markov Chains, this study investigates the data aggregation problem for partially connected networks to obtain: the number of rounds of message exchanges needed to reach a network-convergence, the average convergence rate in a round of message exchange, and the number of rounds required to reach a stationary-convergence.

The Tyranny of the Vital Few: The Pareto Principle in Language Design

Modern high-level programming languages often contain constructs whose semantics are non-trivial. In practice however, software developers generally restrict the use of such constructs to settings in which their semantics is simple (programmers use language constructs in ways they understand and can reason about). As a result, when developing tools for analyzing and manipulating software, a disproportionate amount of effort ends up being spent developing capabilities needed to analyze constructs in settings that are infrequently used. This paper takes the position that such distinctions between theory and practice are an important measure of the analyzability of a language.

Data Aggregation in Static Adhoc Networks

This research is concerned with the data aggregation (DA) algorithms in wireless static networks, where host mobility is low or not provided and the number of nodes is fixed. The DA problem is investigated in the presence of omission faults, which can be a common type of failure in wireless communication. It is shown that, in the worst case, the number of rounds of message-exchange needed to reduce the diameter of values held by the hosts in the network is half the maximum network diameter. The research also obtains the upper bound on the number of rounds of message-exchange to reach the stationary-convergence, i.e. the difference between the agreed upon values among the hosts is within a predefined tolerance.

Adaptive ARQ in wireless sensor networks

currently, in the MAC protocol of wireless sensor network, the sender mote sends data with or without receiving acknowledgements (ACK). Although in some applications, people enable the ACK function, they just use the fixed repeat times. In such scenarios, the sender mote does not care about the quality of the channel when they have to send the data. However, in the real world, the channel may not be good thus resulting in a poor delivery rate. For example, motes that are neighbors of the sender mote may also send data at the same time. This indeed leads to collision. This paper recommends the use of the Adaptive Repeat Request (ARQ) scheme for the MAC with fuzzy logic control to reduce the collisions in the channel thus improving the delivery rate in comparison to the current MAC protocols. We implement our method with only a small number of sensor motes. Obtained results show that ARQ has 5–30% more delivery rate than the other two methods.

ANN based Data Mining Analysis of the Parkinson’s Disease

This paper intends to provide an evaluation of the different pre-processing techniques that can aid a classifier in the classification of the Parkinson Disease (PD) dataset. PD is a chronic and progressive moment disorder caused due to the malfunction and death of vital nerve cells in the brain. The key indications of the chronic malady in the central nervous system can be best captivated from the Mentation, Activities of Daily Life (ADL), Motor Examination, and Complications of Therapy. The speech symptom which is an ADL is a common ground for the progress of the PD. A comprehensive study on the application of different pre-processing techniques is carried out on the PD dataset obtained from the UCI website. For classifying the PD dataset we employed the ANN based MLP classifier. With the objective of improving the prediction accuracies of the healthy, and people with Parkinson disease on the PD dataset this study highlights the fact that the combination of several pre-processing techniques namely Discretization, Resampling, and SMOTE can best aid in the classification process. This study is unique in the sense that we have not come across any similar studies in the Data Mining literature. General Terms Data Mining, Pre-processing techniques, classification techniques.

Exploratory failure analysis of open source software

Reliability growth modeling in software system plays an important role in measuring and controlling software quality during software development. One main approach to reliability growth modeling is based on the statistical correlation of observed failure intensities versus estimated ones by the use of statistical models. Although there are a number of statistical models in the literature, this research concentrates on the following seven models: Weibull, Gamma, S-curve, Exponential, Lognormal, Cubic, and Schneidewind. The failure data collected are from five popular open source software (OSS) products. The objective is to determine which of the seven models best fits the failure data of the selected OSS products as well as predicting the future failure pattern based on partial failure history. The outcome reveals that the best model fitting the failure data is not necessarily the best predictor model.